A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.

Detalhes bibliográficos
Autor(a) principal: KUCK, T. N.
Data de Publicação: 2021
Outros Autores: SANO, E. E., BISPO, P. da C., SHIGUEMORI, E. H., SILVA FILHO, P. B. F., MATRICARDI, E. A. T.
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
Texto Completo: http://www.alice.cnptia.embrapa.br/alice/handle/doc/1136170
Resumo: Abstract: The near-real-time detection of selective logging in tropical forests is essential to support actions for reducing CO2 emissions and for monitoring timber extraction from forest concessions in tropical regions. Current operating systems rely on optical data that are constrained by persistent cloud-cover conditions in tropical regions. Synthetic aperture radar data represent an alternative to this technical constraint. This study aimed to evaluate the performance of three machine learning algorithms applied to multitemporal pairs of COSMO-SkyMed images to detect timber exploitation in a forest concession located in the Jamari National Forest, Rondônia State, Brazilian Amazon. The studied algorithms included random forest (RF), AdaBoost (AB), and multilayer perceptron artificial neural network (MLP-ANN). The geographical coordinates (latitude and longitude) of logged trees and the LiDAR point clouds before and after selective logging were used as ground truths. The best results were obtained when the MLP-ANN was applied with 50 neurons in the hidden layer, using the ReLu activation function and SGD weight optimizer, presenting 88% accuracy both for the pair of images used for training (images acquired in June and October) of the network and in the generalization test, applied on a second dataset (images acquired in January and June). This study showed that X-band SAR images processed by applying machine learning techniques can be accurately used for detecting selective logging activities in the Brazilian Amazon.
id EMBR_be427d3d33af163702908c0e9b38c7dd
oai_identifier_str oai:www.alice.cnptia.embrapa.br:doc/1136170
network_acronym_str EMBR
network_name_str Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
repository_id_str 2154
spelling A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.DesmatamentoSensoriamento RemotoSynthetic aperture radarAbstract: The near-real-time detection of selective logging in tropical forests is essential to support actions for reducing CO2 emissions and for monitoring timber extraction from forest concessions in tropical regions. Current operating systems rely on optical data that are constrained by persistent cloud-cover conditions in tropical regions. Synthetic aperture radar data represent an alternative to this technical constraint. This study aimed to evaluate the performance of three machine learning algorithms applied to multitemporal pairs of COSMO-SkyMed images to detect timber exploitation in a forest concession located in the Jamari National Forest, Rondônia State, Brazilian Amazon. The studied algorithms included random forest (RF), AdaBoost (AB), and multilayer perceptron artificial neural network (MLP-ANN). The geographical coordinates (latitude and longitude) of logged trees and the LiDAR point clouds before and after selective logging were used as ground truths. The best results were obtained when the MLP-ANN was applied with 50 neurons in the hidden layer, using the ReLu activation function and SGD weight optimizer, presenting 88% accuracy both for the pair of images used for training (images acquired in June and October) of the network and in the generalization test, applied on a second dataset (images acquired in January and June). This study showed that X-band SAR images processed by applying machine learning techniques can be accurately used for detecting selective logging activities in the Brazilian Amazon.TAHISA NEITZEL KUCK; EDSON EYJI SANO, CPAC; POLYANNA DA CONCEIÇÃO BISPO; ELCIO HIDEITI SHIGUEMORI; PAULO FERNANDO FERREIRA SILVA FILHO; ERALDO APARECIDO TRONDOLI MATRICARDI.KUCK, T. N.SANO, E. E.BISPO, P. da C.SHIGUEMORI, E. H.SILVA FILHO, P. B. F.MATRICARDI, E. A. T.2021-11-16T18:00:23Z2021-11-16T18:00:23Z2021-11-162021info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleRemote Sensing, v. 13, n. 3341, 2021.http://www.alice.cnptia.embrapa.br/alice/handle/doc/1136170enginfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)instname:Empresa Brasileira de Pesquisa Agropecuária (Embrapa)instacron:EMBRAPA2021-11-16T18:00:34Zoai:www.alice.cnptia.embrapa.br:doc/1136170Repositório InstitucionalPUBhttps://www.alice.cnptia.embrapa.br/oai/requestopendoar:21542021-11-16T18:00:34falseRepositório InstitucionalPUBhttps://www.alice.cnptia.embrapa.br/oai/requestcg-riaa@embrapa.bropendoar:21542021-11-16T18:00:34Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice) - Empresa Brasileira de Pesquisa Agropecuária (Embrapa)false
dc.title.none.fl_str_mv A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
title A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
spellingShingle A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
KUCK, T. N.
Desmatamento
Sensoriamento Remoto
Synthetic aperture radar
title_short A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
title_full A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
title_fullStr A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
title_full_unstemmed A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
title_sort A Comparative Assessment of Machine-Learning Techniques for Forest Degradation Caused by Selective Logging in an Amazon Region Using Multitemporal X-Band SAR Images.
author KUCK, T. N.
author_facet KUCK, T. N.
SANO, E. E.
BISPO, P. da C.
SHIGUEMORI, E. H.
SILVA FILHO, P. B. F.
MATRICARDI, E. A. T.
author_role author
author2 SANO, E. E.
BISPO, P. da C.
SHIGUEMORI, E. H.
SILVA FILHO, P. B. F.
MATRICARDI, E. A. T.
author2_role author
author
author
author
author
dc.contributor.none.fl_str_mv TAHISA NEITZEL KUCK; EDSON EYJI SANO, CPAC; POLYANNA DA CONCEIÇÃO BISPO; ELCIO HIDEITI SHIGUEMORI; PAULO FERNANDO FERREIRA SILVA FILHO; ERALDO APARECIDO TRONDOLI MATRICARDI.
dc.contributor.author.fl_str_mv KUCK, T. N.
SANO, E. E.
BISPO, P. da C.
SHIGUEMORI, E. H.
SILVA FILHO, P. B. F.
MATRICARDI, E. A. T.
dc.subject.por.fl_str_mv Desmatamento
Sensoriamento Remoto
Synthetic aperture radar
topic Desmatamento
Sensoriamento Remoto
Synthetic aperture radar
description Abstract: The near-real-time detection of selective logging in tropical forests is essential to support actions for reducing CO2 emissions and for monitoring timber extraction from forest concessions in tropical regions. Current operating systems rely on optical data that are constrained by persistent cloud-cover conditions in tropical regions. Synthetic aperture radar data represent an alternative to this technical constraint. This study aimed to evaluate the performance of three machine learning algorithms applied to multitemporal pairs of COSMO-SkyMed images to detect timber exploitation in a forest concession located in the Jamari National Forest, Rondônia State, Brazilian Amazon. The studied algorithms included random forest (RF), AdaBoost (AB), and multilayer perceptron artificial neural network (MLP-ANN). The geographical coordinates (latitude and longitude) of logged trees and the LiDAR point clouds before and after selective logging were used as ground truths. The best results were obtained when the MLP-ANN was applied with 50 neurons in the hidden layer, using the ReLu activation function and SGD weight optimizer, presenting 88% accuracy both for the pair of images used for training (images acquired in June and October) of the network and in the generalization test, applied on a second dataset (images acquired in January and June). This study showed that X-band SAR images processed by applying machine learning techniques can be accurately used for detecting selective logging activities in the Brazilian Amazon.
publishDate 2021
dc.date.none.fl_str_mv 2021-11-16T18:00:23Z
2021-11-16T18:00:23Z
2021-11-16
2021
dc.type.driver.fl_str_mv info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv Remote Sensing, v. 13, n. 3341, 2021.
http://www.alice.cnptia.embrapa.br/alice/handle/doc/1136170
identifier_str_mv Remote Sensing, v. 13, n. 3341, 2021.
url http://www.alice.cnptia.embrapa.br/alice/handle/doc/1136170
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.source.none.fl_str_mv reponame:Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
instname:Empresa Brasileira de Pesquisa Agropecuária (Embrapa)
instacron:EMBRAPA
instname_str Empresa Brasileira de Pesquisa Agropecuária (Embrapa)
instacron_str EMBRAPA
institution EMBRAPA
reponame_str Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
collection Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
repository.name.fl_str_mv Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice) - Empresa Brasileira de Pesquisa Agropecuária (Embrapa)
repository.mail.fl_str_mv cg-riaa@embrapa.br
_version_ 1794503512014454784